• Home
  • Services
  • About
  • Contact
  • Blog
  • 知財活動のROICへの貢献
  • 生成AIを活用した知財戦略の策定方法
  • 生成AIとの「壁打ち」で、新たな発明を創出する方法

​
​よろず知財コンサルティングのブログ

GPT-6 Astraより高性能な「社内モデル」

10/9/2026

0 Comments

 
2026年9月8日(現地時間)、OpenAIがミレニアム懸賞問題の一つであるナビエ・ストークス方程式の存在と滑らかさの問題に対する解法を公開しました。数学界では、この成果が本来の懸賞が問う「外力なし」のケースではない点や、第三者による査読が未完了である点等から、評価には慎重です。証明を生成したのは“GPT-6 Astraより大幅に高性能な”社内モデルということで、公開されているモデルより高性能な「社内モデル」が次の高性能な公開モデルを開発するパターンが生まれているようです。
生成AIに、この件について深堀調査を行わせましたので、ご参照ください。なお、生成AIによる調査・分析結果は、公開された情報だけに基づくものであり、必ずしも実情を示したものではなく、誤った情報を含む可能性があることにご留意のうえ、ご参照ください。
 
On the Navier–Stokes Millennium Prize Problem
https://openai.com/index/navier-stokes-solution/
 
 
An “Internal Model” That Outperforms GPT-6 Astra
On September 8, 2026 (local time), OpenAI published a solution to the Navier–Stokes existence and smoothness problem, one of the Millennium Prize Problems. The mathematical community has been cautious in its assessment, however, noting, among other things, that the result does not address the “unforced” case specified in the original prize problem and that independent peer review has not yet been completed. The proof was reportedly generated by an internal model described as “significantly more capable than GPT-6 Astra.” A pattern therefore appears to be emerging in which internal models that outperform publicly available models are used to develop the next generation of more capable models for public release.
I asked generative AI to conduct an in-depth investigation into this topic, and invite you to review the findings. Please bear in mind that the AI-generated research and analysis are based solely on publicly available information, may not necessarily reflect the actual state of affairs, and could contain inaccuracies.

Your browser does not support viewing this document. Click here to download the document.
Your browser does not support viewing this document. Click here to download the document.
Your browser does not support viewing this document. Click here to download the document.
Your browser does not support viewing this document. Click here to download the document.
Your browser does not support viewing this document. Click here to download the document.
Your browser does not support viewing this document. Click here to download the document.
0 Comments



Leave a Reply.

    著者

    萬秀憲

    アーカイブ

    April 2026
    March 2026
    February 2026
    January 2026
    December 2025
    November 2025
    October 2025
    September 2025
    August 2025
    July 2025
    June 2025
    May 2025
    April 2025
    March 2025
    February 2025
    January 2025
    December 2024
    November 2024
    October 2024
    September 2024
    August 2024
    July 2024
    June 2024
    May 2024
    April 2024
    March 2024
    February 2024
    January 2024
    December 2023
    November 2023
    October 2023
    September 2023
    August 2023
    July 2023
    June 2023
    May 2023
    April 2023
    March 2023
    February 2023
    January 2023
    December 2022
    November 2022
    October 2022
    September 2022
    August 2022
    July 2022
    June 2022
    May 2022
    April 2022
    March 2022
    February 2022
    January 2022
    December 2021
    November 2021
    October 2021
    September 2021
    August 2021
    July 2021
    June 2021
    May 2021
    April 2021
    March 2021
    February 2021
    January 2021
    December 2020
    November 2020
    October 2020
    September 2020
    August 2020
    July 2020
    June 2020

    カテゴリー

    All

    RSS Feed

Copyright © よろず知財戦略コンサルティング All Rights Reserved.
サイトはWeeblyにより提供され、お名前.comにより管理されています
  • Home
  • Services
  • About
  • Contact
  • Blog
  • 知財活動のROICへの貢献
  • 生成AIを活用した知財戦略の策定方法
  • 生成AIとの「壁打ち」で、新たな発明を創出する方法